GitHub Actions workflows: parameterize over tests, add summary

Co-authored-by: Benjamin Prevor <[email protected]>

.github/workflows/treadmill-ci-test.yml

@@ -26,7 +26,46 @@ permissions:
   contents: read
 
 jobs:
-  treadmill-ci:
+  analyze-changes:
+    runs-on: ubuntu-latest
+
+    outputs:
+      hwci-tests-json: ${{ steps.analyze-changes.outputs.hwci-tests-json }}
+
+    steps:
+      # This is not run within the context of a repository that contains actual
+      # kernel / userspace code, so there is nothing for us to analyze. Instead
+      # we clone this very repository and select all test definitions:
+      - name: Checkout the tock-hardware-ci repository
+        uses: actions/checkout@v4
+        with:
+          path: tock-hardware-ci
+
+      - name: Checkout the tock/tock repository
+        uses: actions/checkout@v4
+        with:
+          # Checkout the repository at the commit that triggered the workflow
+          repository: tock/tock
+          path: tock-tock
+
+      - name: Select all defined tests
+        id: analyze-changes
+        run: |
+          # Run the select_tests.py script
+          python3 tock-hardware-ci/hwci/select_tests.py \
+            --repo-path tock-tock \
+            --hwci-path tock-hardware-ci/hwci \
+            --output selected_tests.json
+
+          echo "Selected HWCI tests:"
+          cat selected_tests.json
+
+          # Output the tests JSON
+          hwci_tests_json=$(cat selected_tests.json | jq -c '.')
+          echo "hwci-tests-json=${hwci_tests_json}" >> "$GITHUB_OUTPUT"
+
+  run-treadmill-ci:
+    needs: [analyze-changes]
     uses: ./.github/workflows/treadmill-ci.yml
     with:
       # Only run on a specific repository, as others will not have the right
@@ -38,12 +77,14 @@ jobs:
       # appropriate environment (depending on the on: triggers above)
       job-environment: ${{ github.event_name == 'pull_request' && 'treadmill-ci' || 'treadmill-ci-merged' }}
 
+
       # This workflow tests the tock-hardware-ci scripts itself, so take the
       # current GITHUB_SHA:
       tock-hardware-ci-ref: ${{ github.sha }}
 
       # Use the latest upstream Tock kernel / userspace components:
       tock-kernel-ref: 'master'
       libtock-c-ref: 'master'
+      tests-json: ${{ needs.analyze-changes.outputs.hwci-tests-json }}
 
     secrets: inherit

.github/workflows/treadmill-ci.yml

@@ -16,8 +16,10 @@
 # [1]: https://book.treadmill.ci/
 # [2]: https://github.com/treadmill-tb/treadmill
 # [3]: https://book.treadmill.ci/user-guide/github-actions-integration.html
+#
 
 name: treadmill-ci
+
 env:
   TERM: xterm # Makes tput work in actions output
 
@@ -39,13 +41,18 @@ on:
       libtock-c-ref:
         required: true
         type: string
+      tests-json:
+        required: false
+        type: string
+        default: '["tests/c_hello.py"]' # Default to single test for backward compatibility
 
 jobs:
   test-prepare:
     runs-on: ubuntu-latest
 
     # Do not run job on forks, as they will not have the correct environment set up
     if: github.repository == inputs.repository-filter
+
     environment: ${{ inputs.job-environment }}
 
     outputs:
@@ -77,16 +84,6 @@ jobs:
           popd
           echo "$PWD/treadmill/target/debug" >> "$GITHUB_PATH"
 
-      # - uses: actions/checkout@v4
-      #   with:
-      #     repository: tock/tock
-      #     ref: ${{ inputs.tock-kernel-ref }}
-      #     path: tock
-
-      # - name: Analyze changes and determine types of tests to run
-      #   run: |
-      #     echo "TODO: implement this!"
-
       - name: Generate a token to register new just-in-time runners
         id: generate-token
         uses: actions/create-github-app-token@v1
@@ -99,23 +96,21 @@ jobs:
         env:
           GH_TOKEN: ${{ steps.generate-token.outputs.token }}
           TML_API_TOKEN: ${{ secrets.TREADMILL_API_TOKEN }}
-
           # Currently, all tests run only on hosts attached to an nRF52840DK
           DUT_BOARD: nrf52840dk
-
           # A Raspberry Pi OS netboot (NBD) image with a GitHub Actions
           # self-hosted runner pre-configured.
           #
           # For the available images see
           # https://book.treadmill.ci/treadmillci-deployment/images.html
-          IMAGE_ID: 1b6900eff30f37b6d012240f63aa77a22e20934e7f6ebf38e25310552dc08378
-
+          IMAGE_ID: df24da6c7a03d87b1b6b55162383a9dfdf48a129b5f3e648748f0f9d11cdb470
           # Limit the supervisors to hosts that are compatible with this
           # image. This is a hack until we introduce "image sets" which define
           # multiple images for various supervisor hosts, but otherwise behave
           # identically:
           HOST_TYPE: nbd-netboot
           HOST_ARCH: arm64
+          TESTS_JSON: ${{ inputs.tests-json }}
         run: |
           # When we eventually launch tests on multiple hardware platforms in
           # parallel, we need to supply different SUB_TEST_IDs here:
@@ -126,6 +121,7 @@ jobs:
           # runner (connected to the exact board we want to run tests on).
           RUNNER_ID="tml-gh-actions-runner-${GITHUB_REPOSITORY_ID}-${GITHUB_RUN_ID}-${GITHUB_RUN_ATTEMPT}-${SUB_TEST_ID}"
 
+
           # Obtain a new just-in-time runner registration token:
           RUNNER_CONFIG_JSON="$(gh api \
             -H "Accept: application/vnd.github+json" \
@@ -162,28 +158,33 @@ jobs:
           TML_JOB_ID="$(echo "$TML_JOB_ID_JSON" | jq -r .job_id)"
           echo "Enqueued Treadmill job with ID $TML_JOB_ID"
 
+
           # Pass the job IDs and other configuration data into the outputs of
           # this step, such that we can run test-execute job instances for each
           # Treadmill job we've started:
-          echo "tml-job-ids=[ \
-            \"$TML_JOB_ID\" \
-          ]" >> "$GITHUB_OUTPUT"
-
-          echo "tml-jobs={ \
-            \"$TML_JOB_ID\": { \
-              \"runner-id\": \"$RUNNER_ID\", \
-            } \
-          }" >> "$GITHUB_OUTPUT"
+          TML_JOB_IDS_OUTPUT="[ \"$TML_JOB_ID\" ]"
+          echo "Setting tml-job-ids output to ${TML_JOB_IDS_OUTPUT}"
+          echo "tml-job-ids=${TML_JOB_IDS_OUTPUT}" >> "$GITHUB_OUTPUT"
+          TML_JOBS_OUTPUT="{ \"$TML_JOB_ID\": { \"runner-id\": \"$RUNNER_ID\", \"tests\": $TESTS_JSON } }"
+          echo "Setting tml-jobs output to ${TML_JOBS_OUTPUT}"
+          echo "tml-jobs=${TML_JOBS_OUTPUT}" >> "$GITHUB_OUTPUT"
+
+          # Generate an overview over the scheduled jobs and their respective tests:
+          cat <<GITHUB_STEP_SUMMARY >>"$GITHUB_STEP_SUMMARY"
+          ## Test Overview
+          | Test | Board | Job |
+          |------|-------|-----|
+          GITHUB_STEP_SUMMARY
+          echo "$TESTS_JSON" | jq -r -c '.[]' | while read TEST; do
+            echo "| \`$TEST\` | \`$DUT_BOARD\` | [\`$TML_JOB_ID\`](#tml-job-summary-$TML_JOB_ID) |" >>"$GITHUB_STEP_SUMMARY"
+          done
 
   test-execute:
     needs: test-prepare
-
     strategy:
       matrix:
         tml-job-id: ${{ fromJSON(needs.test-prepare.outputs.tml-job-ids) }}
-
     runs-on: ${{ fromJSON(needs.test-prepare.outputs.tml-jobs)[matrix.tml-job-id].runner-id }}
-
     steps:
       - name: Print Treadmill Job Context and Debug Information
         run: |
@@ -207,30 +208,10 @@ jobs:
         run: |
           echo "verbose = off" >> $HOME/.wgetrc
 
-      - name: Checkout the Tock Hardware CI scripts
-        uses: actions/checkout@v4
-        with:
-          repository: tock/tock-hardware-ci
-          ref: ${{ inputs.tock-hardware-ci-ref }}
-
-      - name: Checkout the Tock kernel repository
-        uses: actions/checkout@v4
-        with:
-          path: tock
-          repository: tock/tock
-          ref: ${{ inputs.tock-kernel-ref }}
-
-      - name: Checkout the libtock-c repository
-        uses: actions/checkout@v4
-        with:
-          path: libtock-c
-          repository: tock/libtock-c
-          ref: ${{ inputs.libtock-c-ref }}
-
       - uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
           # Avoid overwriting the RUSTFLAGS environment variable
-          rustflags: ""
+          rustflags: ''
 
       - name: Install required system packages
         run: |
@@ -244,25 +225,113 @@ jobs:
             git cargo openocd python3 python3-pip python3-serial \
             python3-pexpect gcc-arm-none-eabi libnewlib-arm-none-eabi \
             pkg-config libudev-dev cmake libusb-1.0-0-dev udev make \
-            gdb-multiarch gcc-arm-none-eabi build-essential || true
+            gdb-multiarch gcc-arm-none-eabi build-essential jq || true
 
           # Install probe-rs:
           curl --proto '=https' --tlsv1.2 -LsSf \
             https://github.com/probe-rs/probe-rs/releases/latest/download/probe-rs-tools-installer.sh \
             | sh
 
+      - name: Checkout the Tock Hardware CI scripts
+        uses: actions/checkout@v4
+        with:
+          repository: tock/tock-hardware-ci
+          ref: ${{ inputs.tock-hardware-ci-ref }}
+
+      - name: Checkout the Tock kernel repository
+        uses: actions/checkout@v4
+        with:
+          path: tock
+          repository: tock/tock
+          ref: ${{ inputs.tock-kernel-ref }}
+
+      - name: Checkout the libtock-c repository
+        uses: actions/checkout@v4
+        with:
+          path: libtock-c
+          repository: tock/libtock-c
+          ref: ${{ inputs.libtock-c-ref }}
+          fetch-depth: 0  
+          submodules: false
+          persist-credentials: true
+
       - name: Create Python virtual environment and install required dependencies
         run: |
           python3 -m venv ./hwcienv
           source ./hwcienv/bin/activate
           pip install -r hwci/requirements.txt -c hwci/requirements-frozen.txt
 
       - name: Run tests
+        env:
+          JSON_TEST_ARRAY: ${{ toJSON(fromJSON(needs.test-prepare.outputs.tml-jobs)[matrix.tml-job-id].tests) }}
         run: |
           source ./hwcienv/bin/activate
           cd ./hwci
           export PYTHONPATH="$PWD:$PYTHONPATH"
-          python3 core/main.py --board boards/nrf52dk.py --test tests/c_hello.py
+          STEP_FAIL=0
+
+          # Generate a summary of all the tests executed:
+          cat <<GITHUB_STEP_SUMMARY >>"$GITHUB_STEP_SUMMARY"
+          ### <a id="tml-job-summary-${{ matrix.tml-job-id }}"></a>Tests executed on board \`nrf52840dk\`, job ID ${{ matrix.tml-job-id }}
+
+          | Result | Test |
+          |--------|------|
+          GITHUB_STEP_SUMMARY
+
+          while read TEST; do
+          # WHILE ---------->
+
+          # For markdown links, we want a sanitized test name without special characters
+          SANITIZED_TEST_NAME="${TEST//[^[:alnum:]]/_}"
+
+          # Prepare a collapsible section in the test output:
+          cat <<STEP_SUMMARY_DETAILS >>"./step-summary-details.txt"
+          <details>
+          <summary>Test \`$TEST\`</summary>
+          ##### <a id="#test-$SANITIZED_TEST_NAME"></a> Test \`$TEST\`
+
+          \`\`\`
+          STEP_SUMMARY_DETAILS
+
+          # Run the test script, saving its output:
+          echo "===== RUNNING TEST $TEST ====="
+          FAIL=0
+          set -o pipefail
+          python3 core/main.py --board boards/nrf52dk.py --test "$TEST" 2>&1 | tee ./job-output.txt || FAIL=1
+          set +o pipefail
+
+          # Insert the result into the markdown table:
+          if [ "$FAIL" == "0" ]; then
+            echo "| ✅ | \`$TEST\` ([Output](#test-$SANITIZED_TEST_NAME)) |" >>"$GITHUB_STEP_SUMMARY"
+          else
+            echo "===== Test $TEST failed! ====="
+            echo "| ❌ | \`$TEST\` ([Output](#test-$SANITIZED_TEST_NAME)) |" >>"$GITHUB_STEP_SUMMARY"
+            STEP_FAIL=1
+          fi
+
+          # Sanitize the output (remove triple backslashes) and copy it into the step summary details:
+          cat ./job-output.txt | sed 's/```//g' >>"./step-summary-details.txt"
+
+          # Finalize this section of the test output summary:
+          cat <<STEP_SUMMARY_DETAILS >>"./step-summary-details.txt"
+          \`\`\`
+
+          </details>
+
+          STEP_SUMMARY_DETAILS
+
+          # -----------> DONE
+          done < <(echo "$JSON_TEST_ARRAY" | jq -r -c '.[]')
+
+          # Finally, add all test output sections to the overall step summary:
+          echo "" >>"$GITHUB_STEP_SUMMARY"
+          cat "./step-summary-details.txt" >>"$GITHUB_STEP_SUMMARY"
+
+          # Exit with an error if at least one test failed:
+          if [ "$STEP_FAIL" != "0" ]; then
+            echo "At least one test failed, exiting with error."
+            exit 1
+          fi
 
       - name: Request shutdown after successful job completion
         run: |

hwci/select_tests.py

@@ -0,0 +1,57 @@
+# Licensed under the Apache License, Version 2.0 or the MIT License.
+# SPDX-License-Identifier: Apache-2.0 OR MIT
+# Copyright Tock Contributors 2024.
+
+#!/usr/bin/env python3
+
+import os
+import argparse
+import json
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Select all HWCI tests.")
+    parser.add_argument(
+        "--repo-path",
+        type=str,
+        default=".",
+        help="Path to the tock/tock repository to analyze",
+    )
+    parser.add_argument(
+        "--hwci-path",
+        type=str,
+        required=True,
+        help="Path to the tock-hardware-ci repository",
+    )
+    parser.add_argument(
+        "--output",
+        type=str,
+        default="selected_tests.json",
+        help="Output JSON file for selected tests",
+    )
+    args = parser.parse_args()
+
+    # For now, we ignore the repo-path (tock/tock repository) since we are not analyzing changes yet
+    # In the future, we will use repo-path to analyze the changes and select tests accordingly
+
+    # Path to the tests directory within the tock-hardware-ci repository
+    tests_dir = os.path.join(args.hwci_path, "hwci", "tests")
+
+    # Find all .py files in the tests directory
+    test_files = []
+    for root, dirs, files in os.walk(tests_dir):
+        for file in files:
+            if file.endswith(".py"):
+                # Get the relative path to the test file
+                test_path = os.path.relpath(os.path.join(root, file), args.hwci_path)
+                test_files.append(test_path)
+
+    # Output the list of test files as a JSON array
+    with open(args.output, "w") as f:
+        json.dump(test_files, f)
+
+    print(f"Selected HWCI tests: {test_files}")
+
+
+if __name__ == "__main__":
+    main()